Qwen AI Chat

Qwen is Alibaba’s AI model family for writing in several languages, structured analysis and code review. Start chatting with Qwen 3.8 Flash for free, then sign up to save your chats and get Qwen 3.8 Max on Pro.

What Qwen is good at

Qwen is good at bilingual customer messages, side-by-side business comparisons and finding edge-case bugs in code. In our tests, Qwen 3.8 Max found the empty-array bug in a JavaScript average function, wrote matching English and Spanish customer emails and compared launch options in a table with its assumptions stated.

Multilingual writing

Write a friendly order-delay email in English and Spanish. The parcel is two days late. Offer a refund of shipping costs, without promising a new delivery date.

Structured analysis

Compare launching a product in one country first versus three countries at once. Use a table covering cost, learning speed and operational risk. State your assumptions.

Code review

Review this JavaScript function for bugs and give a corrected version: function average(xs) { return xs.reduce((a, b) => a + b) / xs.length; } Explain the empty-array case.

In the email sample, Qwen said the shipping refund had already been issued, although the prompt only asked it to offer one. Three sample prompts tested live by Chandan Kumar on November 4, 2026. Answers are unedited and regenerated for every new model.

Qwen models and pricing

Qwen 3.8 Flash costs the least, from 1 Sparks per message. Qwen 3.8 Max costs the most, from 10.

ModelSpeedReasoningContextSparksPlanEst. messages a month on Pro
Qwen 3.8 Flash LatestFastBasic991KSparks1+Free≈10,000
Qwen 3.8 Max StandardAdvanced262.1KSparks10+Pro≈1,000
Qwen 3.7 Flash FastBasic991KSparks1+Free≈10,000
Qwen 3.7 Plus StandardBasic1MSparks5+Pro≈2,000
Qwen 3.7 Max StandardAdvanced991KSparks5+Pro≈2,000
Qwen 3.6 Plus StandardBasic1MSparks5+Pro≈2,000
Qwen 3.5 Flash FastBasic1MSparks1+Free≈10,000

Sparks are per message and start at the amount shown. Monthly counts are estimates; longer chats and attached files use more.

All plans

Qwen 3.8 Max specs

Context window

262.1K
tokens

The largest on Sparkian is 1.1M tokens. Attached files count toward it.

Sparks per message

Sparks10+

Included with the Pro plan.

Est. messages a month on Pro

≈1,000

$19 a month with 10,000 Sparks included.

Inputs
Text, Images
Outputs
Text, Code
Web tools
Search and read websites
Released
August 2, 2026
Access
Pro
Speed
Standard
Reasoning
Advanced

Qwen release dates

Qwen 3.8 Flash is the newest Qwen model, released August 26, 2026. You can use all 7 versions on Sparkian.

  1. Qwen 3.8 Flash Latest

    Context window
    991K tokens
    Sparks
    Sparks1+
    Access
    Free
  2. Qwen 3.8 Max

    Context window
    262.1K tokens
    Sparks
    Sparks10+
    Access
    Pro
  3. Qwen 3.7 Flash

    Context window
    991K tokens
    Sparks
    Sparks1+
    Access
    Free
  4. Qwen 3.7 Plus

    Context window
    1M tokens
    Sparks
    Sparks5+
    Access
    Pro
  5. Qwen 3.7 Max

    Context window
    991K tokens
    Sparks
    Sparks5+
    Access
    Pro
  6. Qwen 3.6 Plus

    Context window
    1M tokens
    Sparks
    Sparks5+
    Access
    Pro
  7. Qwen 3.5 Flash

    Context window
    1M tokens
    Sparks
    Sparks1+
    Access
    Free

Why use Qwen on Sparkian

  • Compare Qwen with Claude Haiku and DeepSeek

    See how Qwen holds up against other models in one prompt.

  • One plan, 100+ models

    No separate subscription per provider.

  • Research with sources

    Web search and reading, cited.

  • Images and teams

    Generate in chat, share projects.

100+ models from more than 20 AI providers, all on one plan.

Qwen questions, answered.

Pricing, plans and which Qwen model to pick.

Written by
Chandan Kumar
Updated
November 11, 2026

Qwen is a trademark of Alibaba. Sparkian is not affiliated with or endorsed by model providers.

Ready to spark your next idea?

Your next question, first draft, or big idea. Start here.